Robert O Keohane-After Hegemony_ Cooperation and Discord in the World Political Economy-Princeton Univ Pr (1984)

Vista previa del material en texto

AFTER HEGEMONY
This Page Intentionally Left Blank
AFTER HEGEMONY
Cooperation and Discord in the
World Political Economy
ROBERT O. KEOHANE
Princeton University Press
Princeton, New Jersey
Copyright © 1984 by Princeton University Press
Published by Princeton University Press, 41 William Street,
Princeton, New Jersey 08540
In the United Kingdom: Princeton University Press,
Chichester, West Sussex
All Rights Reserved
Library of Congress Cataloging-in-Publication Data
Keohane, Robert O. (Robert Owen), 1941-
After hegemony.
Bibliography: p.
Includes index.
1. International economic relations. 2. World politics
—1945- . I. Title.
HF1411.K442 1984 337 84-42576
ISBN 0-691-07676-6 
This book has been composed in Linotron Sabon
To NannerI Overholser Keohane
This Page Intentionally Left Blank
PREFACE
In its genesis and support, this is an old-fashioned book. It is essentially
the work of an individual scholar, unaided by a research team or large-
scale funding. Nevertheless, I have accumulated a number of institu-
tional debts of gratitude during the seven years of research and writing.
I benefited, during the early stages of reflection and reading, from
being a Fellow of the Center for Advanced Study in the Behavioral
Sciences during 1977-78, under a grant from the German Marshall
Fund of the United States. Most of the research was done while I was
teaching at Stanford University until the spring of 1981 and at Brandeis
University since then. Stanford helped me to finance research assistance
and a trip to the International Energy Agency in Paris in 1981. The
Mazur Fund for Faculty Research at Brandeis supplied funds for pho-
tocopying the manuscript and circulating it to colleagues. Thanks to
a sabbatical leave generously provided by Brandeis University for the
academic year 1983-84, I was able to devote myself wholeheartedly,
between June 1983 and January 1984, to preparing the final manu-
script. Wellesley College permitted me to use its convenient and well-
organized library and to take advantage of its computer system for
word-processing, which greatly expedited my work. Staff members of
both the library and the computer center were most helpful. For all
of this support I am most grateful.
The overall argument of this book has never appeared in print
before, although ''The demand for international regimes," published
in International Organization, Spring 1982, contains early versions of
some of the core ideas of chapters 5-6. The theme of Part III—the
complementarity of hegemony and cooperation in practice—is also
first presented here, but some of the case material has been published
before. Chapter 8 builds on "Hegemonic leadership and U.S. foreign
economic policy in the 'Long Decade' of the 1950s," published in
William P. Avery and David P. Rapkin, eds., America in a Changing
World Political Economy (New York: Longman, 1982). Chapter 9 is
in part based on "The theory of hegemonic stability and changes in
international economic regimes, 1967-1977." Sections of this chapter
that reproduce parts of the earlier article, in modified form, are re-
printed by permission of Westview Press from Ole R. Holsti, Randolph
M. Siverson, and Alexander L. George, eds., Change in the Interna-
tional System (copyright 1980 by Westview Press, Boulder, Colorado).
Some of chapter 10 also appeared in "International agencies and the
vii
PREFACE
art of the possible: the case of the IEA," Journal of Policy Analysis
and Management, vol. 1, no. 4 (Summer 1982), copyright 1982 by
John Wiley & Sons.
This book on cooperation also benefited from the noninstitutional
cooperation of scores of friends, students, and colleagues—so many,
indeed, that I refrain from trying to list them all, lest some inadvertently
be excluded. Earlier versions of several chapters, in draft or as pre-
viously published articles, were circulated to quite a few political sci-
entists and economists, and I received many helpful observations, all
of which I seriously considered and many of which led to changes.
The willingness of scholars to devote time and intelligence to helping
each other improve the quality of their work is one of the most re-
warding features of contemporary academic life. Fortunately for me,
the field of international political economy contains a large number
of very talented and generous people.
I do want to mention by name a small number of people who have
made special contributions. Karen Bernstein and Shannon Salmon
served ably as research assistants, gathering material used in chapters
8-10. I shared many early, otherwise uncirculated drafts with Helen
Milner. I am grateful to her both for offering trenchant criticisms and
for not giving up on the project even when my preliminary arguments
may have seemed hopelessly contorted and confused. Joseph Nye, my
close friend and former co-author, has been a valuable source of both
intellectual perspective and moral support. Vinod Aggarwal, Robert
Axelrod, James Caporaso, Benjamin Cohen, Robert Gilpin, Peter
Gourevitch, Leah Haus, Harold Jacobson, Peter Katzenstein, Nannerl
Keohane, David Laitin, Helen Milner, Joseph Nye, Susan Moller Okin,
Robert Putnam, and Howard Silverman read all or large parts of the
penultimate draft and gave me valuable comments.
Equally important are the senior scholars whom I have sought to
emulate: creative people who respect and care about younger thinkers,
and who refuse to hide self-protectively behind reputations and titles.
These intellectuals are willing to propose new ideas and to submit
them to scrutiny. Knowing that social science advances not so much
by the cumulative grubbing of facts as by the dialectical confrontation
of ideas, they are not afraid to be criticized or even proven wrong.
Among these mentors I include particularly Alexander George, Ernst
Haas, Albert Hirschman, Stanley Hoffmann, Charles Kindleberger,
Robert North, Raymond Vernon, and Kenneth Waltz—a diverse set
of scholars united only by their fertile imaginations, intellectual hon-
esty, and vigor of mind and spirit.
Two men who were inspirations to me are no longer alive. One is
viii
PREFACE
Fred Hirsch, an imaginative political economist, author of Social Lim-
its to Growth, and a great person who died too young. The other is
Robert E. Keohane, my father. Although he had a powerful intellect,
he never produced major works of scholarship; but my memory of
the range and richness of his knowledge and his utter integrity still
serve to warn me against superficiality and opportunism.
The remaining members of my immediate family have made major
contributions to this enterprise. My mother, Mary P. Keohane, has
for over forty years provided me with a synergistic combination of
maternal love, moral precepts, and intellectual stimulation. She con-
tinues to be supporter, critic, and exemplar to me. Concern for my
children's futures reinforces my belief in the urgency of understanding
cooperation in world politics; but they themselves have more often
reminded me that sometimes scholarship should be subordinated to
fun. Nannerl Overholser Keohane, my wife, has played such an enor-
mously important and multifaceted role that it is difficult for me to
convey its significance. Her own writings set a high standard for depth
of research, clarity of expression, and grace of style. Her accomplish-
ments as college president both fill me with admiration and bolster
my determination to make the most of the happy life of scholarship
that dedicated people in such positions make possible. Her criticisms
of my works and her high expectations for them impel me to greater
levels of effort. In addition to everything else, she has been a source
of love, moral support, and domestic contentment.
Wellesley, Massachusetts
January 1984
ix
This Page Intentionally Left Blank
CONTENTSPREFACE vii
I • Questions and Concepts
1 • Realism, Institutionalism, and Cooperation 5
2 • Politics, Economics, and the International System 18
3 • Hegemony in the World Political Economy 31
II • Theories of Cooperation and International Regimes
4 • Cooperation and International Regimes 49
5 • Rational-Choice and Functional Explanations 65
6 • A Functional Theory of International Regimes 85
7 • Bounded Rationality and Redefinitions of
Self-interest 110
III • Hegemony and Cooperation in Practice
8 • Hegemonic Cooperation in the Postwar Era 135
9 • The Incomplete Decline of Hegemonic Regimes 182
10 • The Consumers' Oil Regime, 1974-81 217
IV • Conclusion
11 • The Value of Institutions and the Costs
of Flexibility 243
BIBLIOGRAPHY 260
INDEX 281
This Page Intentionally Left Blank
AFTER HEGEMONY
This Page Intentionally Left Blank
PART I
Questions and Concepts
This Page Intentionally Left Blank
• 1 •
REALISM, INSTITUTIONALISM,
AND COOPERATION
Since the repeal of the "iron law of wages," economics has ceased to
be the "dismal science." Economists no longer believe that most people
must exist at the subsistence level, but argue, on the contrary, that
gradual improvement in the material conditions of human life is pos-
sible. Yet while economics has become more cheerful, politics has
become gloomier. The twentieth century has seen an enormous ex-
pansion of real and potential international violence. In the world po-
litical economy, opportunities for conflict among governments have
increased as the scope of state action has widened. The greatest dangers
for the world economy, as well as for world peace, have their sources
in political conflicts among nations.
In the study of politics, perhaps nothing seems so dismal as writing
about international cooperation. Indeed, when I told a friend and
former teacher of mine that I was writing a book on this subject, she
replied that it would have to be a short book. Was I planning extra-
large type and wide margins to justify hard covers?
I could have retorted that my book would also discuss discord, a
much more common feature of world politics. Yet the issue goes deeper
than that. International cooperation among the advanced industrial-
ized countries since the end of World War II has probably been more
extensive than international cooperation among major states during
any period of comparable length in history. Certainly the extent and
complexity of efforts to coordinate state economic policies have been
greater than they were between the two world wars, or in the century
before 1914. Yet cooperation remains scarce relative to discord be-
cause the rapid growth of international economic interdependence
since 1945, and the increasing involvement of governments in the
operation of modern capitalist economies, have created more points
of potential friction. Interdependence can transmit bad influences as
well as good ones: unemployment or inflation can be exported as well
as growth and prosperity. American steel workers may lose their jobs
because of subsidies to European steel producers by the European
Economic Community and European governments; high interest rates
in the United States may constrain economic activity abroad.
Interdependence leads democratic governments to expand state ac-
5
QUESTIONS AND CONCEPTS
tivity in order to protect their citizens from fluctuations in the world
economy (Cameron, 1978). When this state activity takes the form of
seeking to force the costs of adjustment onto foreigners, international
discord results. Thus even a rising absolute level of cooperation may
be overwhelmed by discord, as increased interdependence and gov-
ernmental intervention create more opportunities for policy conflict.
As in Alice in Wonderland, it may be necessary to keep running faster
in order to stand still. Scholars should not wait for cooperation to
become the rule rather than the exception before studying it, for ig-
norance of how to promote cooperation can lead to discord, conflict,
and economic disaster before cooperation ever has a chance to prevail.
This book is about how cooperation has been, and can be, organized
in the world political economy when common interests exist. It does
not concentrate on the question of how fundamental common interests
can be created among states. Thus two topics that could legitimately
be treated in a book on international economic cooperation are not
systematically considered: I neither explore how economic conditions
affect patterns of interests, nor do I investigate the effects of ideas and
ideals on state behavior. The theory that I develop takes the existence
of mutual interests as given and examines the conditions under which
they will lead to cooperation. I begin with the premise that even where
common interests exist, cooperation often fails. My purpose is to
diagnose the reasons for such failure, and for the occasional successes,
in the hope of improving our ability to prescribe remedies.
Because I begin with acknowledged common interests, my study
focuses on relations among the advanced market-economy countries,
where such interests are manifold. These countries hold views about
the proper operation of their economies that are relatively similar—
at least in comparison with the differences that exist between them
and most less developed countries, or the nonmarket planned econ-
omies. They are engaged in extensive relationships of interdependence
with one another; in general, their governments' policies reflect the
belief that they benefit from these ties. Furthermore, they are on friendly
political terms; thus political-military conflicts between them compli-
cate the politics of economic transactions less than they do in East-
West relations.
The arguments of this book surely apply to some relationships be-
tween the advanced market-economy countries and less developed
countries. These states have interests in common, which can only be
realized through cooperation. To perhaps a more limited extent, my
analysis should also be relevant to those areas of East-West relations
where common interests exist. The focus of this book on cooperation
6
REALISM AND INSTITUTIONALISM
among the advanced industrialized countries by no means implies that
cooperation is impossible, or unnecessary, between North and South
or East and West. To illustrate and test my ideas about cooperation
and discord, however, I focus first on the area where common interests
are greatest and where the benefits of international cooperation may
be easiest to realize. Careful extension of this argument into East-
West and North-South relations, including security as well as eco-
nomic issues, would be most welcome.
REALISM, INSTITUTIONALISM, AND COOPERATION
Impressed with the difficulties of cooperation, observers have often
compared world politics to a "state of war." In this conception, in-
ternational politics is "a competition of units in the kind of state of
nature that knows no restraints other than those which the changing
necessities of the game and the shallow conveniences of the players
impose" (Hoffmann, 1965, p. vii). It is anarchic in the sense that it
lacks an authoritative government that can enact and enforce rules of
behavior. States must rely on "the means they can generate and the
arrangements they can make for themselves" (Waltz, 1979, p. 111).
Conflict and war result, since each state is judge in its own cause and
can use force to carry out its judgments (Waltz, 1959, p. 159). The
discord that prevails is accounted for by fundamental conflicts of
interest (Waltz, 1959; Tucker, 1977).
Were this portrayal of world politics correct, any cooperation that
occurs would be derivative from overall patterns of conflict. Alliance
cooperation would be easy to explain as a result of the operation of
a balanceof power, but system-wide patterns of cooperation that
benefit many countries without being tied to an alliance system directed
against an adversary would not. If international politics were a state
of war, institutionalized patterns of cooperation on the basis of shared
purposes should not exist except as part of a larger struggle for power.
The extensive patterns of international agreement that we observe on
issues as diverse as trade, financial relations, health, telecommunica-
tions, and environmental protection would be absent.
At the other extreme from these "Realists" are writers who see co-
operation as essential in a world of economic interdependence, and
who argue that shared economic interests create a demand for inter-
national institutions and rules (Mitrany, 1975). Such an approach,
which I refer to as "Institutionalist" because of its adherents' emphasis
on the functions performed by international institutions, runs the risk
of being naive about power and conflict. Too often its proponents
7
QUESTIONS AND CONCEPTS
incorporate in their theories excessively optimistic assumptions about
the role of ideals in world politics, or about the ability of statesmen
to learn what the theorist considers the "right lessons." But sophis-
ticated students of institutions and rules have a good deal to teach us.
They view institutions not simply as formal organizations with head-
quarters buildings and specialized staffs, but more broadly as "rec-
ognized patterns of practice around which expectations converge"
(Young, 1980, p. 337). They regard these patterns of practice as sig-
nificant because they affect state behavior. Sophisticated institution-
alists do not expect cooperation always to prevail, but they are aware
of the malleability of interests and they argue that interdependence
creates interests in cooperation.1
During the first twenty years or so after World War II, these views,
though very different in their intellectual origins and their broader
implications about human society, made similar predictions about the
world political economy, and particularly about the subject of this
book, the political economy of the advanced market-economy coun-
tries. Institutionalists expected successful cooperation in one field to
"spill over" into others (Haas, 1958). Realists anticipated a relatively
stable international economic order as a result of the dominance of
the United States. Neither set of observers was surprised by what
happened, although they interpreted events differently.
Institutionalists could interpret the liberal international arrange-
ments for trade and international finance as responses to the need for
policy coordination created by the fact of interdependence. These ar-
rangements, which we will call "international regimes," contained
rules, norms, principles, and decisionmaking procedures. Realists could
reply that these regimes were constructed on the basis of principles
espoused by the United States, and that American power was essential
1
 In a preliminary draft I referred to "Functionalists" rather than "Institutionalists,"
since the scholars to whom I am alluding often adopted the former label or some variant
of it, for themselves. On the suggestion of a reader, however, I altered the terminology
in order to avoid confusion between "Functionalism" and the functional theory of
international regimes presented in chapter 6. It should be emphasized that, as noted in
the text, I employ a stylized contrast between Realism and Institutionalism to focus
sharply on the issues addressed by this book, not to identify any given author with a
simplistic variant of either position. For instance, although Stanley Hoffmann writes of
international relations as "a state of war," his highly nuanced view of world politics
would not normally be considered representative of Realism. Among the Institutionalists
as well, there is substantial variation. Ernst Haas, for instance, has taken state power
more seriously, and has been more cautious about the growth of international insti-
tutions, than David Mitrany.
8
REALISM AND INSTITUTIONALISM
for their construction and maintenance. For Realists, in other words,
the early postwar regimes rested on the political hegemony of the
United States. Thus Realists and Institutionalists could both regard
early postwar developments as supporting their theories.
After the mid-1960s, however, U.S. dominance in the world political
economy was challenged by the economic recovery and increasing
unity of Europe and by the rapid economic growth of Japan. Yet
economic interdependence continued to grow, and the pace of in-
creased U.S. involvement in the world economy even accelerated after
1970. At this point, therefore, the Institutionalist and Realist predic-
tions began to diverge. From a strict Institutionalist standpoint, the
increasing need for coordination of policy, created by interdependence,
should have led to more cooperation. From a Realist perspective, by
contrast, the diffusion of power should have undermined the ability
of anyone to create order.
On the surface, the Realists would seem to have made the better
forecast. Since the late 1960s there have been signs of decline in the
extent and efficacy of efforts to cooperate in the world political econ-
omy. As American power eroded, so did international regimes. The
erosion of these regimes after World War II certainly refutes a naive
version of the Institutionalist faith in interdependence as a solvent of
conflict and a creator of cooperation. But it does not prove that only
the Realist emphasis on power as a creator of order is valid. It might
be possible, after the decline of hegemonic regimes, for more sym-
metrical patterns of cooperation to evolve after a transitional period
of discord. Indeed, the persistence of attempts at cooperation during
the 1970s suggests that the decline of hegemony does not necessarily
sound cooperation's death knell.
International cooperation and discord thus remain puzzling. Under
what conditions can independent countries cooperate in the world
political economy? In particular, can cooperation take place without
hegemony and, if so, how? This book is designed to help us find
answers to these questions. I begin with Realist insights about the role
of power and the effects of hegemony. But my central arguments draw
more on the Institutionalist tradition, arguing that cooperation can
under some conditions develop on the basis of complementary inter-
ests, and that institutions, broadly defined, affect the patterns of co-
operation that emerge.
Hegemonic leadership is unlikely to be revived in this century for
the United States or any other country. Hegemonic powers have his-
torically only emerged after world wars; during peacetime, weaker
countries have tended to gain on the hegemon rather than vice versa
9
QUESTIONS AND CONCEPTS
(Gilpin, 1981). It is difficult to believe that world civilization, much
less a complex international economy, would survive such a war in
the nuclear age. Certainly no prosperous hegemonic power is likely
to emerge from such a cataclysm. As long as a world political economy
persists, therefore, its central political dilemma will be how to organize
cooperation without hegemony.
COOPERATION AND VALUES
Cooperation is elusive enough, and its sources are sufficiently multi-
faceted and intertwined, that it constitutes a difficult subject to study.
It is particularly hard, perhaps impossible, to investigate with scientific
rigor. No sensible person would choose it as a topic of investigation
on the grounds that its puzzles could readily be "solved." I study it,
despite the lack of rich, multi-case data suitable for the testing of
hypotheses and despite the relative paucity of relevant theory, because
of its normative significance.
This choice poses problems both for the author andfor the reader.
My values necessarily affect my argument; yet I am sufficiently pos-
itivistic to attempt to distinguish between my empirical and normative
assertions. Except for this chapter and chapter 11, After Hegemony
represents an attempt at theoretical, historical, and interpretive anal-
ysis rather than an exercise in applied ethics. I seek to increase our
understanding of cooperation, in the belief that increased understand-
ing can help to improve political amity and economic welfare, though
not with the naive supposition that knowledge necessarily increases
either amity or welfare. I try to provide an account of cooperation
that can be analyzed, if not tested in a strict sense, by others who do
not share my normative views, even as I recognize that, except for my
own values, I would never have decided to write this book. Yet since
I can surely not keep my analysis entirely distinct from my values, it
seems fair to the reader for me to indicate briefly my thoughts about
whether, or under what conditions, international cooperation is a
"good" that we should strive to increase.
Cooperation is viewed by policymakers less as an end in itself than
a means to a variety of other objectives. To inquire about the moral
value of cooperation is partly to ask about the ends for which it is
pursued. Along with many others, I would disapprove of cooperation
among the governments of wealthy, powerful states to exploit poorer,
weaker countries. Even if the goals sought through cooperation were
judged desirable in principle, particular attempts to achieve them could
have perverse effects. That is, the consequences of cooperation could
10
REALISM AND INSTITUTIONALISM
be adverse, either for certain countries not fully represented in decision-
making or for overall world welfare. When the conventional inter-
national economic wisdom is misguided, cooperation can be worse
than doing nothing. So the economic orthodoxy of 1933 appeared to
Franklin Delano Roosevelt when he wrecked the London Economic
Conference of that year (Feis, 1966); and so does the internationally
oriented Keynesianism of the Carter Administration now appear to
economic theorists of rational expectations who put their trust in
markets (Saxonhouse, 1982). Under conditions of interdependence,
some cooperation is a necessary condition for achieving optimal levels
of welfare; but it is not sufficient, and more cooperation may not
necessarily be better than less.
Although it would be naive to believe that increased cooperation,
among any group of states for whatever purposes, will necessarily
foster humane values in world politics, it seems clear that more ef-
fective coordination of policy among governments would often help.
Internationally minded Keynesians recommend extensive harmoni-
zation of macroeconomic policies (Whitman, 1979). Even proponents
of international laissez-faire, who reject these proposals, have to rec-
ognize that free markets depend on the prior establishment of property
rights (North and Thomas, 1973; Field, 1981; Conybeare, 1980; North,
1981). People may disagree on what forms of international cooper-
ation are desirable and what purposes they should serve, but we can
all agree that a world without any cooperation would be dismal indeed.
In the conclusion, I return explicitly to the problem of moral eval-
uation. Is it good that the international regimes discussed in this book
exist? In what ways are they deficient when evaluated by appropriate
moral standards? Would it have been better had they never come into
being? No comprehensive or definitive answers to these questions are
offered, but the importance of the problem of ethical evaluation de-
mands that they be raised.
THE PLAN OF THIS BOOK
I hope that After Hegemony will be read not only by students of world
politics but also by economists interested in the political underpinnings
of the international economy and by citizens concerned about inter-
national cooperation. To encourage readers outside of political science,
I have tried to eliminate professional jargon wherever possible and to
define my terms clearly using ordinary language. Yet since this book
is meant for people with different disciplinary backgrounds, and since
it draws on disparate traditions to do so, its key concepts may be
11
QUESTIONS AND CONCEPTS
easily misunderstood. I hope that readers will be careful not to seize
on words and phrases out of context as clues to pigeonholing my
argument. Is it "liberal" because I discuss cooperation, or "mercan-
tilist" because I emphasize the role of power and the impact of he-
gemony? Am I a "radical" because I take Marxian concepts seriously,
or a "conservative" because I talk about order? The simplemindedness
of such inferences should be obvious.
Since I use concepts from economics to develop a political theory
about cooperation and discord in the world economy, I need to be
particularly clear about my definitions of economics and politics and
my conception of theory. Chapter 2 discusses these questions, as a
necessary prologue to the development of my theory in Part II. Chapter
3 then prepares the ground for a serious analysis of cooperation, and
the effects of institutions on it, by examining the "theory of hegemonic
stability," which holds that order, in the Realist lexicon, depends on
the preponderance of a single state. Chapter 3 argues that although
hegemony can facilitate cooperation, it is neither a necessary nor a
sufficient condition for it. We will see later that hegemony is less
important for the continuation of cooperation, once begun, than for
its creation.
Part II, which constitutes the theoretical core of this book, begins
by defining two key terms, "cooperation" and "international regimes."
Since these terms are used in chapter 3 before their full elaboration
in chapter 4, it is important to note here that cooperation is defined
in a deliberately unconventional way. Cooperation is contrasted with
discord; but is also distinguished from harmony. Cooperation, as com-
pared to harmony, requires active attempts to adjust policies to meet
the demands of others. That is, not only does it depend on shared
interests, but it emerges from a pattern of discord or potential discord.
Without discord, there would be no cooperation, only harmony.
It is important to define cooperation as mutual adjustment rather
than to view it simply as reflecting a situation in which common
interests outweigh conflicting ones. In other words, we need to dis-
tinguish between cooperation and the mere fact of common interests.
We require this distinction because discord sometimes prevails even
when common interests exist. Since common interests are sometimes
associated with cooperation but sometimes with discord, cooperation
is evidently not a simple function of interests. Especially where un-
certainty is great and actors have different access to information, ob-
stacles to collective action and strategic calculations may prevent them
from realizing their mutual interests. The mere existence of common
12
REALISM AND INSTITUTIONALISM
interests is not enough: institutions that reduce uncertainty and limit
asymmetries in information must also exist.
Using chapter 4's definitions of cooperation and international re-
gimes, chapters 5-7 present my functional theory of international re-
gimes. Chapter 5 employs game theory and collective goods theory to
argue that "the emergence of cooperation among egoists" (Axelrod,
1981, 1984) is possible, even in the absence of common government,
but that the extent of such cooperation will depend on the existence
of international institutions, or international regimes, with particular
characteristics. Rational-choice theory enables us to demonstrate that
the pessimistic conclusions about cooperation often associated with
Realism are not necessarily valid, even if we acceptthe assumption of
rational egoism. Chapter 6 then uses theories of market failure in
economics, as well as more conventional rational-choice theory, to
develop a functional theory of international regimes that shows why
governments may construct regimes and even abide by their rules.
According to this argument, regimes contribute to cooperation not by
implementing rules that states must follow, but by changing the con-
text within which states make decisions based on self-interests. Inter-
national regimes are valuable to governments not because they enforce
binding rules on others (they do not), but because they render it pos-
sible for governments to enter into mutually beneficial agreements with
one another. They empower governments rather than shackling them.
Chapter 7 relaxes our earlier assumptions of rationality and narrow
egoism. First it explores the implications of deviating from the premise
of classic rationality by assuming, more realistically, that decisions are
costly for governments to make. That is, governments operate under
the constraints of "bounded rationality" (Simon, 1955), rather than
as classically rational actors. On this assumption, regimes do not sub-
stitute for continuous calculations of self-interest (which are impos-
sible), but rather provide rules of thumb to which other governments
also adhere. These rules may provide opportunities for governments
to bind their successors, as well as to make other governments' policies
more predictable. Cooperation fostered by awareness of bounded ra-
tionality does not require that states accept common ideals or renounce
fundamental principles of sovereignty. Even egoistic actors may agree
to accept obligations that preclude making calculations about advan-
tage in particular situations, if they believe that doing so will have
better consequences in the long run than failure to accept any rules
or acceptance of any other politically feasible set of rules.
Chapters 5-6 and the first two sections of chapter 7 adopt the
premise of egoism. The last two sections of chapter 7 relax this as-
13
QUESTIONS AND CONCEPTS
sumption by distinguishing between egoistic self-interest and concep-
tions of self-interest in which empathy plays a role. Actors that inter-
pret their interests as empathetically interdependent, in our terminology,
may find it easier to form international regimes than those whose
definitions of self-interest are more constricted. I explore the strengths
and limitations of egoist and empathetic interpretations of state be-
havior by analyzing two features of the world political economy that
may appear puzzling from an egoistic standpoint: the facts that regime
rules and principles are sometimes treated as having morally obligatory
status and that unbalanced exchanges of resources often persist over
a considerable period of time.
The argument of Part II, taken as a whole, constitutes both a critique
and modification of Realism. Realist theories that seek to predict in-
ternational behavior on the basis of interests and power alone are
important but insufficient for an understanding of world politics. They
need to be supplemented, though not replaced, by theories stressing
the importance of international institutions. Even if we fully under-
stand patterns of power and interests, the behavior of states (and of
transnational actors as well) may not be fully explicable without un-
derstanding the institutional context of action.
This institutionalist modification of Realism provides some rather
abstract answers to the major puzzle addressed by this book: namely,
how can cooperation take place in world politics in the absence of
hegemony? We understand the creation of regimes as a result of a
combination of the distribution of power, shared interests, and pre-
vailing expectations and practices. Regimes arise against the back-
ground of earlier attempts, successful or not, at cooperation. Fur-
thermore, the theory of Part II explains the continuation of existing
regimes even after the conditions that facilitated their creation have
disappeared: regimes acquire value for states because they perform
important functions and because they are difficult to create or recon-
struct. In order to realize fully the significance of this theoretical ar-
gument for understanding contemporary international regimes, we
need to combine it with a historical understanding of the creation of
contemporary international regimes and of their evolution since the
end of World War II. This is the task of Part III.
Part III argues that the creation of contemporary international re-
gimes can largely be explained by postwar U.S. policy, implemented
through the exercise of American power. As American economic pre-
ponderance eroded between the 1950s and the 1970s, major inter-
national economic regimes came under pressure. Thus far Realist ex-
pectations are met. Yet the changes in these regimes did not always
14
REALISM AND INSTITUTIONALISM
correspond to the shifts in power, and the decline of American he-
gemony did not lead uniformly to the collapse of regimes. Cooperation
persists and, on some issues, has increased. Current patterns of discord
and cooperation reflect interacting forces: the remaining elements of
American hegemony as well as the effects of its erosion, the current
mixture of shared and conflicting interests, and the international eco-
nomic regimes that represent an institutional legacy of hegemony.
The first step in the empirical analysis of Part III is to examine how
American hegemony actually operated. Chapter 8 therefore discusses
American hegemony during the two decades of U.S. dominance, span-
ning the years from the Truman Doctrine and Marshall Plan (1947)
to the late 1960s, when the United States began to show signs of
seeking to protect itself from the impact of economic interdependence.
The sources and practices of hegemonic cooperation are the focus of
attention here. The episodes studied in this chapter illustrate the in-
timate connection between discord and cooperation pointed out in
chapter 4, and they also reveal that inequalities of power can be quite
consistent with mutual adjustment, policy coordination, and the for-
mation of international regimes. Hegemony and international regimes
may be complementary, or even to some extent substitute for each
other: both serve to make agreements possible and to facilitate com-
pliance with rules.
This period of hegemonic cooperation was short: Henry Luce's
"American Century" was under severe pressure after less than twenty
years. No system-level theory accounts for this, since—as chapter 8
shows—one of the most important reasons for the brevity of U.S.
dominance was the pluralistic nature of American politics.2 Given a
decline in American power, however, believers in the theory of heg-
emonic stability would predict a decline in cooperation. Chapter 9
evaluates the applicability of this theory to the evolution of interna-
tional regimes for money, trade, and oil between the mid-1960s and
the early 1980s. Did international regimes embodying patterns of heg-
emonic cooperation become less effective because of the erosion of
American power? Chapter 9 shows that the pattern of regime change
varied a great deal from one issue-area to another, and that shifts in
American power were of different significance in international finance,
trade, and oil. The decline of American hegemony provides only part
2
 The question of the causes of the decline of American hegemony, not addressed
systematically in this volume, is dealt with in an original and provocative way by Robert
Gilpin (1975, 1981).
15
QUESTIONS AND CONCEPTS
of the explanation for the decline of postwar international regimes.
Chapter 9 reaches this conclusion without itself attempting to provide
a complete explanation for this phenomenon, since to do so wouldrequire close examination of the effects of changes in macroeconomic
conditions and international economic competitiveness, the role of
ideas and learning, and the impact of domestic politics on foreign
economic policies in the United States and elsewhere.
Chapter 9 also points out that although international regimes came
under pressure in the 1970s, the advanced industrialized countries
continue to coordinate their policies, albeit imperfectly, on interna-
tional economic issues. Contemporary attempts to cooperate reflect
not only the erosion of hegemony but the continued existence of in-
ternational regimes, most of which had their origins in American he-
gemony. Old patterns of cooperation work less well than they did,
partly because U.S. hegemony has declined; but the survival of patterns
of mutual policy adjustment, and even their extension, can be facili-
tated by international regimes that had their origins in the period of
hegemony. From chapter 9 we can see that both Realist concepts of
power and self-interest and the arguments developed here about the
significance of international regimes provide valuable insights into the
nature of the contemporary world political economy. We need to go
beyond Realism, not discard it. By documenting the erosion of Amer-
ican hegemony, furthermore, chapter 9 shows that our key puzzle—
how cooperation can take place in the absence of hegemony—is not
merely hypothetical but highly topical. It thus suggests the relevance
for our own era of the theories of international cooperation put for-
ward in Part II. Shared interests and existing institutions make it
possible to cooperate, but the erosion of American hegemony makes
it necessary to do so in new ways.
Chapter 10 further explores how regimes affect patterns of coop-
eration by investigating in detail the most significant international
economic regime established among the advanced industrialized coun-
tries since 1971: the energy arrangements, revolving around the In-
ternational Energy Agency (IEA), set up under U.S. leadership after
the oil crisis of 1973-74. This regime was not global, but was limited
to oil-consuming countries and competed with another partial regime
formed by oil producers. We will see in chapter 10 that regime-oriented
efforts at cooperation do not always succeed, as the fiasco of IEA
actions in 1979 illustrates, but that they can have a positive impact
under relatively favorable conditions, as the events of 1980 suggest.
Chapter 10 also lends some support to the general proposition that
successful attempts to use international regimes to facilitate cooper-
16
REALISM AND INSTITUTIONALISM
ation depend on efforts to reduce the costs of transactions involved
in policy coordination and on measures to provide information to
governments, rather than on enforcement of rules.
The final chapter reviews the argument as a whole, assesses the
moral value of cooperation, and examines implications for policy. My
discussion of ethics concludes that, despite some defects in their
principles, contemporary international regimes are morally acceptable,
at least conditionally. They are easy to justify on the basis of criteria
that stress the importance of the autonomy of states, although eval-
uation is more difficult when cosmopolitan and egalitarian standards
are employed. The policy implications of the book stem most directly
from my emphasis on the value of information produced and distrib-
uted by international regimes. Providers as well as recipients of in-
formation benefit from its availability. It can therefore make sense to
accept obligations that restrain one's own freedom of action in un-
known future situations if others also accept responsibilities, since the
effect of these reciprocal actions is to reduce uncertainty. Assumptions
about the value of "keeping one's options open" therefore need to be
rethought. The pursuit of flexibility can be self-defeating: like Ulysses,
it may be better, on occasion, to have oneself tied to the mast.
17
• 2•
POLITICS, ECONOMICS, AND
THE INTERNATIONAL
SYSTEM
Robert Gilpin has offered a helpful working definition of the phrase
"world political economy" (1975, p. 43):
In brief, political economy in this study means the reciprocal and
dynamic interaction in international relations of the pursuit of
wealth and the pursuit of power.
Causality is reciprocal rather than unidirectional: on the one hand,
the distribution of power creates patterns of property rights within
which wealth is produced and distributed; on the other hand, changes
in productive efficiency and access to resources affect relations of
power in the long term. The interaction between wealth and power is
dynamic because both wealth and power are continually altered, as
are the connections between them.
Wealth and power are linked in international relations through the
activities of independent actors, the most important of which are states,
not subordinated to a worldwide governmental hierarchy. There is no
authoritative allocator of resources: we cannot talk about a "world
society" making decisions about economic outcomes. No consistent
and enforceable set of comprehensive rules exists. If actors are to
improve their welfare through coordinating their policies, they must
do so through bargaining rather than by invoking central direction.
In world politics, uncertainty is rife, making agreements is difficult,
and no secure barriers prevent military and security questions from
impinging on economic affairs. In addition, disagreements about how
benefits should be distributed permeate the relations among actors and
persist because bargains are never permanently valid. Actors are con-
tinually tempted to try imposing burdens on others rather than ab-
sorbing costs of adjustment themselves. Furthermore, this struggle to
make others adjust is played repeatedly. Apparent victory can be il-
lusory or defeat ephemeral, for political bargaining and maneuver
result not in definitive choices conferring power on some people rather
than others, but in agreements that may in the future be reversed or
in discord that signals a continuation of bargaining and maneuver.
All of this is understood by students of international relations. More
18
POLITICS AND ECONOMICS
difficult to grasp are the meanings of the basic and misleadingly fa-
miliar terms "wealth" and "power." Gilpin (1975, p.23) defines wealth
as "anything (capital, land, or labor) that can generate future income;
it is composed of physical assets and human capital (including em-
bodied knowledge)." The problem with this definition is that it seems
to limit wealth to investment goods, excluding assets that merely pro-
vide value in consumption. Edible foodstuffs, ample quantities of gas-
oline for pleasure driving, and decorous jewelry are all, in ordinary
language, considered to be wealth; but they would be excluded by
Gilpin's definition. Adam Smith's definition of wealth as "the annual
produce of the land and labour of a society" (1776/1976, p. 4) avoids
this difficulty, but creates another one, since it refers to a flow of
income rather than a stock of assets. Our ordinary contemporary usage
of "wealth" refers to a stock rather than a flow concept. Taking this
into account, we could follow Karl Polanyi, regarding wealth as "the
means of material want satisfaction" (1957/1971, p. 243). Polanyi's
definition, however, is also subject to telling objections. Lionel Robbins
(1932, p. 9) pointed out a half-century ago that if economics refers
to the satisfaction of material wants alone, it includes the services of
the cook but not of the dancer. Yet although the cook produces a
material product, the ultimate end (the pleasure of eating tasty food)
may be quite as immaterial as the ultimate purpose of attending the
ballet or the opera.
Considering this objection, we could define wealthsimply as the
"means of want satisfaction," or anything that yields utility, whether
in the form of investment or consumption. This definition has the
virtue of viewing wealth as a stock of resources, without arbitrarily
excluding either consumption goods or nonmaterial sources of satis-
faction. Yet it remains excessively broad in two ways. First, it omits
reference to scarcity. In neoclassical economic analysis, value is de-
rivative from market relationships: wealth can only be assessed after
markets have evaluated different products or services. What is not
scarce has no market value. Pure water, for instance, might be con-
sidered a "produce of the land," but in an ecologically pristine society,
it would not constitute wealth because it would be free. In neoclassical
value theory, "exchange value" rather than "use value" is decisive.
Second, even if we take scarcity into account, we still need to distin-
guish between valued experiences that cannot be exchanged for money
without altering their intrinsic nature (such as love, acts of pure friend-
ship, and the ability to make others feel that they are in a state of
grace) and those that can be so exchanged (such as sexual acts with
strangers, favors done for business associates, and the ability to pro-
19
QUESTIONS AND CONCEPTS
duce "that Pepsi feeling"). This is done by limiting "wealth" to means
of want satisfaction that are not only scarce but also marketable: they
can be bought and sold on a market. Thus "the pursuit of wealth" in
the world political economy refers to the pursuit of marketable means
of want satisfaction, whether these are to be used for investment or
consumed by their possessors.
Gilpin asserts that "the nature of power is even more elusive than
that of wealth." Rather than enter the "intradisciplinary squabble"
about it, however, he follows Hans J. Morgenthau's definition of
power as "man's control over the minds and actions of other men."
Power, for Gilpin, refers to a causal relationship and varies according
to the context in which it is exercised: "there is no single hierarchy
of power in international relations" (1975, p. 24).
To define power in terms of control is plausible enough, but it does
not address the question of the value of the concept in the study of
politics. To use the concept of power to explain behavior, one needs
to be able to measure power prior to the actions being explained and
to construct models in which different amounts or types of power lead
to different outcomes. What James G. March has called "basic force
models" are designed to achieve this purpose by using tangible meas-
ures of power resources—such as numbers of people, quality of weap-
ons, or wealth—to predict outcomes of political contention. Yet the
predictions of these models are inaccurate, partly because some actors
care more about particular outcomes than others and are therefore
willing to use greater proportions of their resources to attain them
(March, 1966; Harsanyi, 1962/1971). Thus basic force models, such
as the "crude" theory of hegemonic stability discussed in chapter 3,
are only useful as first approximations. These models can be qualified
by adding auxiliary hypotheses that refer to the role of intangible
factors such as will, intensity of motivation, or—in the "refined" ver-
sion of the theory of hegemonic stability—leadership. Unfortunately,
however, these factors can only be measured after the event. Power is
no longer used to account for behavior; rather, it provides a language
for describing political action.
We saw above that, in the neoclassical economic theory of value,
wealth is not used as a primary category to explain demand or prices;
on the contrary, value (hence wealth) is inferred from demand and
supply, as reflected in price movements on markets. Thus the concepts
of power and wealth have a common deficiency as the basis for ex-
planations of behavior: to estimate the power of actors, or whether a
given product, service, or raw material constitutes wealth, one has to
observe behavior—in power relationships or in markets. To use the
20
POLITICS AND ECONOMICS
Figure 2.1. Politics and Economics: A Schematization
Wealth
(To what extent are wants satisfied through
production and exchange of marketable goods and services?)
High Medium Low
High "pure
Power (What role Political politics"
does control over Medium Economy
other actors ploy
in the process of Low "pure (pure love?)
concepts of power and wealth to account for the behavior that allows
us to identify their presence would be to engage in circular reasoning.
Thus the insight expressed by Gilpin, that the world political economy
revolves around power and wealth, does not enable us to construct
strong explanations of the behavior that we observe.
Nevertheless, defining international political economy as the recip-
rocal and dynamic interaction of the pursuit of wealth and the pursuit
of power is useful from a descriptive standpoint. We can view inter-
national political economy as the intersection of the substantive area
studied by economics—production and exchange of marketable means
of want satisfaction—with the process by which power is exercised
that is central to politics. Wherever, in the economy, actors exert power
over one another, the economy is political. This area of intersection
can be contrasted with "pure economics," in which no actor has any
control over others but faces an externally determined environment.
One can also imagine a situation, also an "ideal type," in which
noneconomic resources were used solely in pursuit of values that could
not be exchanged on a market, such as status, or power itself. Such
a situation would be one of "pure politics." Figure 2.1 makes these
points schematically.
As figure 2.1 illustrates, attempts to separate a sphere of real activity,
called "economics," from another sphere of real activity, called "pol-
itics," are doomed to frustration and failure. Very little of the polity
in modern societies is untouched by the economy, and vice versa; even
apart from questions of governmental intervention, much of the mod-
ern economy is political because firms, unions, and other organizations
seek to exert control over one another. Defined in purely economic or
political terms, the world economy and the international political sys-
21
want sattisfaction?) economics (mysticism?)
QUESTIONS AND CONCEPTS
tern are both abstractions. In the real world of international relations,
most significant issues are simultaneously political and economic.
We have seen that thinking about international political economy
in terms of wealth and power does not enable us to construct strong
explanatory models of behavior. Yet focusing on the pursuit of wealth
and power does contribute to insightful interpretation, since it provides
us with working hypotheses about the motivations of actors that em-
phasize specific interests rather than ideology or rhetoric. To remind
ourselves about wealth and power is a useful antidote to a onesided
emphasis on interdependence and the problem of realizing common
interests. In reflecting on the later discussion in this volume of inter-
national regimes, the reader should keep in mind that these regimes
are rarely if ever instituted by disinterested idealists for the sake of
the common good. Instead, they are constructed principally by gov-
ernments whose officials seek to further the interests of their states (as
they interpret them) and of themselves. They seek wealth and power,
and perhaps other values as well, no matter how much they may
indulge in rhetoric about global welfare or a world "safe for inter-
dependence."
THE COMPLEMENTARITY OF WEALTH AND POWER
Reflection on wealth and power as state objectives soon yields the
conclusion that they are complementary. For contemporary statesmen,
as for the mercantilists of theseventeenth and eighteenth centuries,
power is a necessary condition for plenty, and vice versa. Two ex-
amples, considered in more detail in later chapters, illustrate the point.
In the late 1940s American power was used to build international
economic arrangements consistent with the structure of American cap-
italism; conversely, U.S. military strength depended in the long run
on close economic as well as political ties between the United States
on the one hand and Western Europe and Japan on the other. To say
that American economic or political goals were primary, as historians
involved in controversies over the Cold War often do, is to miss the
point, which is that U.S. economic interests abroad depended on es-
tablishing a political environment in which capitalism could flourish,
and that American political and security interests depended on eco-
nomic recovery in Europe and Japan. The two sets of objectives were
inextricably linked, and similar policies were required to achieve them.
Likewise, when the United States proposed the establishment of an
international energy agency in 1974 to help cope with the shift of
power over oil to producing countries, it did so both to deal with the
22
POLITICS AND ECONOMICS
economic consequences of higher oil prices and to reinforce its own
political influence. Effective international action to alleviate economic
distress seemed impossible without American leadership; conversely,
U.S. influence and prestige were likely to be enhanced by leading a
successful collective effort to ensure energy security.
The complementarity of wealth and power provides a thread of
continuity between the world political economy of the seventeenth
century and that of today. Most governments still appear to adhere
to the propositions that Jacob Viner ascribes to seventeenth-century
mercantilists (1948, p. 10):
1) Wealth is an absolutely essential means to power, whether for
security or for aggression; 2) power is essential or valuable as a
means to the acquisition or retention of wealth; 3) wealth and
power are each proper ultimate ends of national policy; 4) there
is a long-run harmony between these ends, although in particular
circumstances it may be necessary for a time to make economic
sacrifices in the interest of military security and therefore also of
long-run prosperity.
The qualification Viner offers to his fourth point is important. In the
short run, tradeoffs exist between the pursuit of power and the pursuit
of wealth. One of the tasks of students of international political econ-
omy is to analyze these tradeoffs, without forgetting the long-run
complementarity underlying them.
The key tradeoffs for the United States in the 1980s, as for mer-
cantilist statesmen in the seventeenth century and American leaders
in the late 1940s, are not between power and wealth but between the
long-term power/wealth interests of the state and the partial interests
of individual merchants, workers, or manufacturers on the one hand
or short-term interests of the society on the other. The United States
is not the only country that has been unable to formulate long-term
goals without making concessions to partial economic interests. Viner
observes that in Holland during the seventeenth and eighteenth cen-
turies, "where the merchants to a large extent shared directly in gov-
ernment, major political considerations, including the very safety of
the country or its success in wars in which it was actually participating,
had repeatedly to give way to the cupidity of the merchants and their
reluctance to contribute adequately to military finance" (1948, p. 20).
In Britain also, "the autonomy of business connections and traditions,"
according to Viner, hindered the pursuit of state interests. During the
Marshall Plan years, American administrators had to deal with "the
special demands of the American business and agricultural community
23
QUESTIONS AND CONCEPTS
that expected direct and early profit from the program—and who were
well-represented in Congress. ... The general goals of multilateral
trade were certainly in the interests of all these constituencies; yet they,
unlike the State Department, were willing to undermine the achieve-
ment of the general aim for even the smallest immediate gain" (Kolko
and Kolko, 1972, pp. 444-45).
The conflict between short-run and long-run objectives arises largely
in the form of choices between consumption on the one hand and
savings or investment on the other. When the economy underinvests,
it is favoring the present over the future. One can use similar concepts
in discussing power. A state invests in power resources when it binds
allies to itself or creates international regimes in which it plays a central
role. During the 1930s Germany followed a "power approach" to
trade questions, changing the structure of foreign trade so that its
partners would be vulnerable to its own actions (Hirschman, 1945/
1980). After World War II American policy had a broader geograph-
ical focus and was less coercive, but it also stressed power investment.
The United States absorbed short-run economic costs, such as those
imposed by discrimination in the early 1950s against American goods
in Europe, for the sake of political influence that could lead to longer-
term gains. It established international regimes that revolved around
Washington, and on which its allies were highly dependent.
Power disinvestment may also take place; power can be consumed
and not replaced. Governments may be able to maintain levels of
consumption in the present by running current account deficits, bor-
rowing abroad to compensate for low levels of saving at home, as the
United States did during the first few years of the 1980s. In the long
run, however, such policies are unsustainable and erode the bases of
influence, or creditworthiness, on which they depend.
Whether to invest in additional power resources or to consume some
of those that have been accumulated is a perennial issue of foreign
policy. Many of the most important choices governments face have
to do with the relative weight given to consumption (of wealth or
power) versus investment, and with devising strategies for action that
are both viable in the short run and capable of achieving wealth and
power objectives in the long run. Any analysis of the world political
economy must keep in mind the extent to which investments, in power
as well as production, are being made or dissipated. Some of these
investments will be reflected in international regimes and the leadership
strategies that help to construct and maintain them. Defining inter-
national political economy in terms of the pursuit of wealth and power
24
POLITICS AND ECONOMICS
leads us to analyze cooperation in the world political economy less as
an effort to implement high ideals than as a means of attaining self-
interested economic and political goals.
SYSTEMIC ANALYSIS OF INTERNATIONAL POLITICS
Wealth and power are sought by a variety of actors in world politics,
including nonstate organizations such as multinational business cor-
porations (Keohane and Nye, 1972). But states are crucial actors, not
only seeking wealth and power directly but striving to construct frame-
works of rules and practices that will enable them to secure these
objectives, among others, in the future. Our analysis of international
cooperation and regimes therefore focuses principally on states.
State behavior can be studied from the "inside-out" or from the
"outside-in" (Waltz, 1979, p. 63). "Inside-out," or unit-level, expla-
nations locate the sources of behavior within the actor—for instance,
in a country's political or economic system, the attributes of its leaders,
or its domestic political culture. "Outside-in," or systemic, explana-
tions account for state behavior on the basis of attributes of the system
as a whole. Any theorywill, of course, take into account the distinctive
characteristics of actors as well as of the system itself. But a systemic
theory regards these internal attributes as constants rather than as
variables. The variables of a systemic theory are situational: they refer
to the location of each actor relative to others (Waltz, 1979, pp. 67-
73; Keohane, 1983, p. 508). Systemic analysis of the international
political economy begins by locating actors along the dimension of
relative power on the one hand and wealth on the other.
Kenneth Waltz has convincingly shown the error of theorizing at
the unit level without first reflecting on the effects of the international
system as a whole. There are two principal reasons for this. First,
causal analysis is difficult at the unit level because of the apparent
importance of idiosyncratic factors, ranging from the personality of a
leader to the peculiarities of a given country's institutions. Parsimo-
nious theory, even as a partial "first cut," becomes impossible if one
starts analysis here, amidst a confusing plethora of seemingly relevant
facts. Second, analyzing state behavior from "inside-out" alone leads
observers to ignore the context of action: the pressures exerted on all
states by the competition among them. Practices such as seeking to
balance the power of potential adversaries may be accounted for on
the basis of distinctive characteristics of the governments in question
when they could be explained more satisfactorily on the basis of en-
25
QUESTIONS AND CONCEPTS
during features of world politics. Without prior systemic theory, unit-
level analysis of world politics floats in an empirical and conceptual
vacuum (Waltz, 1979, chs. 4-5).
For these reasons, the analysis of this book begins at the systemic
level. I focus on the effects of system characteristics because I believe
that the behavior of states, as well as of other actors, is strongly affected
by the constraints and incentives provided by the international envi-
ronment. When the international system changes, so will incentives
and behavior. My "outside-in" perspective is therefore similar to that
of systemic forms of Realist theory, or "structural Realism" (Krasner,
1983). What distinguishes my argument from structural Realism is my
emphasis on the effects of international institutions and practices on
state behavior. The distribution of power, stressed by Realists, is surely
important. So is the distribution of wealth. But human activity at the
international level also exerts significant effects. International regimes
alter the information available to governments and the opportunities
open to them; commitments made to support such institutions can
only be broken at a cost to reputation. International regimes therefore
change the calculations of advantage that governments make. To try
to understand state behavior simply by combining the structural Real-
ist theory based on distribution of power and wealth with the foreign
policy analyst's stress on choice, without understanding international
regimes, would be like trying to account for competition and collusion
among oligopolistic business firms without bothering to ascertain
whether their leaders met together regularly, whether they belonged
to the same trade associations, or whether they had developed informal
means of coordinating behavior without direct communication. In-
ternational regimes not only deserve systematic study; they virtually
cry out for it.
Yet no systemic analysis can be complete. When we come to our
discussion of the postwar international political economy in Part III,
we will have to look beyond the system toward accounts of state
behavior that emphasize the effects of domestic institutions and lead-
ership on patterns of state behavior. That is, we will have to introduce
some unit-level analysis as well. In doing so, we will pay special at-
tention to the most powerful actor in the world political economy,
the United States. Since the United States shaped the system as much
as the system shaped it, and since it retained greater leeway for au-
tonomous action than other countries throughout the thirty-five years
after World War II, we have to look at the United States from the
inside-out as well as from the outside-in.
26
POLITICS AND ECONOMICS
LIMITATIONS OF SYSTEMIC ANALYSIS
My choice of systemic theory as a place to begin analysis does not
imply that I regard it as completely satisfactory even as a "first cut."
Before going on to the systemic analysis of Part II, therefore, it is
necessary to indicate some of its limitations.
The prevailing model for systemic analysis in politics comes from
economics—in particular, from microeconomic theory. Such theory
posits the existence of firms, with given utility functions (such as profit
maximization), and attempts to explain their behavior on the basis of
environmental factors such as the competitiveness of markets. It is
systemic rather than unit-level theory because its propositions depend
on variations in attributes of the system, not of the units (Waltz, 1979,
pp. 89-91, 93-95, 98). Firms are assumed to act as rational egoists.
Rationality means that they have consistent, ordered preferences, and
that they calculate costs and benefits of alternative courses of action
in order to maximize their utility in view of those preferences. Egoism
means that their utility functions are independent of one another: they
do not gain or lose utility simply because of the gains or losses of
others. Making these assumptions means that rationality and concep-
tions of self-interest are constants rather than variables in systemic
theory. Variations in firms' behavior are accounted for not by varia-
tions in their values, or in the efficiency of their internal organizational
arrangements, but by variations in characteristics of the economic
system—for instance, whether its market structure is competitive, oli-
gopolistic, or monopolistic. Without the assumptions of egoism and
rationality, variations in firms' behavior might have to be accounted
for by differences in values or in their calculating, choice-making abil-
ities. In that case, analysis would revert to the unit level, and the
parsimony of systemic theory—resting on only a small number of
variables—would be lost.1
Systemic theories based on rational-egoist assumptions work best
when there is one uniquely superior course of action. Arnold Wolfers
pointed out this feature of such theories long ago, in arguing that they
provide the best predictions when there is extreme "compulsion," as
in the case of a fire breaking out in a house that has only one exit.
For such a situation, "decision-making analysis would be useful only
in regard to individuals who decided to remain where they were rather
than join the general and expected rush" (1962, p. 14). Spiro J. Latsis
(1976) has more recently argued, in similar terms, that microeconomic
1
 For a fully developed version of this argument, see Keohane, 1983.
27
QUESTIONS AND CONCEPTS
theory based on rational-choice assumptions performs best when ap-
plied to "single-exit" situations. Under these conditions, what Latsis
calls the research program of "situational determinism" works very
well. We do not need to understand the idiosyncracies of the actors
to explain their behavior, since the situation they face mandates that
they must act in a particular way. They will do so if they are rational;
if they fail to do so, they may (if the environmental conditions are
stringent) cease to exist.
This research program has had great success in situations of pure
competition or pure monopoly—and, by extension, in situations that
approximate these ideal types. Situational determinism works under
these circumstances because there is no power competition in pure
competition or pure monopoly. Either economic actors adjust their
behavior tosignals from an impersonal market (in competition), or
they dominate the market (in monopoly). In neither case do they have
to react to the actions of others. As Latsis puts it (1976, pp. 25-26):
Under perfect competition entrepreneurs do not really compete
with each other. The situation may be compared to that of a
player in an n-person game where n is very large. Such games are
reducible to one-person games against nature where the opponent
has no objectives and no known strategy. The "nature" of perfect
competition is unusually strict in allowing a choice between fol-
lowing a single strategy or going under.
Pure monopoly, usually regarded as the exact opposite of per-
fect competition is in fact its heuristic twin. ... The monopolist
maximizes on the basis of his knowledge of the market conditions
and the application of the simple optimizing rule. As with perfect
competition, so with monopoly the "rational" decision-maker
will arrive at the uniquely determined optimal decision by a simple
calculation.
Difficulties arise for this research program under conditions of oli-
gopoly, or "monopolistic competition." Under these conditions, the
situation can be treated as a variable-sum game, played repeatedly
over an indefinite period of time, with a small number of players. This
type of game does not have a determinate solution for any actor,
independent of the behavior of others. It is a "multiple-exit" situation,
and arbitrary assumptions are required to reach unique solutions to
it (Latsis, 1976, pp. 26-39). As we will see in chapters 5 and 6, rational-
egoist calculations of whether to cooperate with one another under
these conditions will depend heavily on the expectations of actors
about others' behavior—and therefore on the nature of institutions.
28
POLITICS AND ECONOMICS
Microeconomic theory does not generate precise, accurate predictions
about behavior in situations of strategic interdependence (Simon,
1976). And as we have seen, strategic interdependence, which bedevils
only part of economics, afflicts the entire study of international politics.
CONCLUSIONS
Systemic analysis will not yield determinate predictions about states'
pursuit of wealth and power. Even if it did, these predictions would
be subject to inaccuracy insofar as great variations in state behavior
resulted from variations in their internal characteristics. Nevertheless,
systemic theory can help us understand how the constraints under
which governments act in the world political economy affect their
behavior. As in Cournot models of oligopoly, however, we also need
to be able to specify something about actors' "reaction functions"—
how they will respond to others' behavior (Fellner, 1949). To do this
on the basis of empirical information, rather than arbitrarily, we must
investigate the institutional context, including the "cues" provided to
actors by rules, practices, and informal patterns of action. That is, we
are led from strictly power-based, game-theoretic analysis toward the
study of international regimes.
Admittedly, accepting rational-egoist assumptions involves taking
seriously a purely hypothetical notion of rationality that does not
accurately model actual processes of human choice (McKeown, 1983b).
Yet beginning with assumptions of egoism and rationality has three
important virtues. First, it simplifies our premises, making deductions
clearer. Second, it directs our attention toward the constraints imposed
by a system on its actors, since it holds the internal determinants of
choice constant. This helps to retain our focus on systemic con-
straints—whether the result of unequal distributions of power or wealth
in the world or of international institutions and practices—rather than
on domestic politics. Finally, adopting the assumption of rational ego-
ism places the argument of this book on the same foundation as that
of Realist theories. The argument here for the importance of inter-
national regimes does not depend on smuggling in assumptions about
altruism or irrationality. Starting with similar premises about moti-
vations, I seek to show that Realism's pessimism about welfare-in-
creasing cooperation is exaggerated. Having done this, in chapter 7
I relax the assumption of classical rationality and the assumption of
egoistic, independent utility maximization, to see how the theory of
regime functions developed earlier on rational-egoist grounds is af-
fected by these changes in premises.
29
QUESTIONS AND CONCEPTS
My interest in both the structure of world power and the institutions
and practices devised by human beings reflects a concern with con-
straints and choices in world politics. The constraints imposed by
distributions of wealth and power are often severe. As Marx said, we
make our own history not just as we please, but "under circumstances
directly found, given, and transmitted from the past" (1852/1972, p.
437). Yet since we do make our own history, there is some room for
choice at any point in time, and over a period of time some of the
constraints can themselves be altered. The limitations of deterministic
theory along the lines of nineteenth-century physics may be disturbing
for social scientists who still carry obsolete images of natural science
in their heads ("weighing like a nightmare on the brains of the living,"
to use another Marxian phrase). But they offer hope for policy. They
suggest that human beings may be able to learn: to develop institutions
and practices that will enable them to cooperate more effectively with-
out renouncing the pursuit of self-interest. The weakness of theory,
but the hope for policy, lies in the fact that people adapt their strategies
to reality. This book seeks to show how adaptive strategies of insti-
tution-building can also change reality, thereby fostering mutually
beneficial cooperation.
30
• 3•
HEGEMONY IN THE WORLD
POLITICAL ECONOMY
It is common today for troubled supporters of liberal capitalism to
look back with nostalgia on British preponderance in the nineteenth
century and American dominance after World War II. Those eras are
imagined to be simpler ones in which a single power, possessing su-
periority of economic and military resources, implemented a plan for
international order based on its interests and its vision of the world.
As Robert Gilpin has expressed it, "the Pax Britannica and Pax Amer-
icana, like the Pax Romana, ensured an international system of relative
peace and security. Great Britain and the United States created and
enforced the rules of a liberal international economic order" (1981,
p. 144).
Underlying this statement is one of the two central propositions of
the theory of hegemonic stability (Keohane, 1980): that order in world
politics is typically created by a single dominant power. Since regimes
constitute elements of an international order, this implies that the
formation of international regimes normally depends on hegemony.
The other major tenet of the theory of hegemonic stability is that the
maintenance of order requires continued hegemony. As Charles P.
Kindleberger has said, "for the world economy to be stabilized, there
has to be a stabilizer, one stabilizer" (1973, p. 305). This implies that
cooperation, which we define in the next chapter as mutual adjustment
of state policies to one another, also depends on the perpetuation of
hegemony.
I discuss hegemony before elaborating my definitions of cooperation
and regimes because my emphasis on how international institutions
such as regimes facilitate cooperation only makes sense if cooperation
and discord are not determined simply by interests and power. In this
chapter I argue that a deterministic version of the theory of hegemonic
stability, relying only on the Realist concepts of interests and power,
is indeed incorrect. There is some validity in a modest version of the
first proposition of the